Paper 14085-43
Paper 14085-43
From RGB to hyperspectral: a deep learning pipeline approach for accessible biomedical sample analysis
16 April 2026 • 15:10 - 15:30 CEST | Luxembourg/Salon 2 (Niveau/Level 0)
Abstract
Hyperspectral imaging (HSI) is a powerful non-invasive technique that combines imaging and spectroscopy to capture detailed spectral information across multiple wavelengths. However, HSI systems are typically complex and expensive, factors that constrain their widespread adoption. This project presents the development of a multispectral imaging (MSI) prototype employing time-multiplexed LED illumination across 31 wavelengths spanning 400–700 nm. The system leverages deep learning algorithms, specifically convolutional neural networks (CNNs), to reconstruct hyperspectral data from multispectral images. This approach offers significant advantages, including low cost, high mobility, high frame rates, and efficient spectral reconstruction. The end-to-end deep learning model demonstrates the ability to learn complex patterns and achieve high accuracy in spectral reconstruction, making it suitable for medical imaging, remote sensing, and other fields requiring detailed spectral analysis.
Presenter
Universidade do Estado do Amazonas (Brazil)
Moisés Oliveira dos Santos is a Professor of Physics at the State University of Amazonas in northern Brazil. In 2012, he earned his PhD from the Institute of Energy and Nuclear Research - IPEN, where he completed a doctoral dissertation titled "Ablation of burned skin with ultrashort pulsed laser to promote healing: Evaluation by optical coherence tomography, histology, microATR-FTIR, and non-linear microscopy."
He has extensive experience in biophotonics, primarily focusing on the analysis of hyperspectral images of biological tissue samples in the mid-infrared range, as well as other techniques, including OCT, SHG, TPEFM, and histology. In recent years, he has expanded his research into data analysis using machine learning and deep learning, integrating these approaches into image and spectral analysis.
Outside of his work with "bio things," his time is dedicated to his wife, Ticiane, and their wonderful, atypical son, Guilherme, who loves Pokémon, and to training in Muay Thai.